ESTRO 35 Abstract-book

S260

ESTRO 35 2016

_____________________________________________________________________________________________________

Medisch Spectrum Twente, Radiotherapy, Enschede, The

Netherlands

VSL, VSL, Delft, The Netherlands

University Medical Centre Utrecht, Department of

Radiotherapyy, Utrecht, The Netherlands

Purpose or Objective:

To independently validate patient-

specific quality assurance (QA) methods, clinically used in the

Netherlands, for IMRT and VMAT plans using the same set of

treatment plans for all institutes.

Material and Methods:

A set of treatment plans was devised:

simple and more complex IMRT/VMAT and a stereotactic

VMAT plan, all 6MV for both Varian and Elekta linacs. Ten

plans were used for Varian linacs (5 for True Beam and 5 for

Clinac) and 9 for Elekta linac(4 for MLCi and 5 for Agility).

The plans were imported in the participating institute’s

treatment planning system for dose computation on the CT

scan of the audit phantom (provided by the audit team

together with the plans). Additionally, 10x10 cm2 fields were

made and computed on both phantoms. Next, the audit team

performed measurements using the audit equipment. All 21

Dutch radiotherapy institutes were audited. The

measurements were performed using an ionization chamber

(PinPoint, PTW), Gafchromic EBT3 film and a 2D ionization

chamber array, all in an octagonal phantom (Octavius, PTW).

Differences between the measured and computed dose

distribution were investigated using a global gamma analysis

with a 5%/1mm criterion for the stereotactic VMAT plan and

3%/3mm for the other plans with a 95% pass rate tolerance.

Additionally, the participating centres performed QA

measurements of the same treatment plans according to

their local protocol and equipment.

Results:

The average difference between the point

measurement, at the centre of the phantom, and the planned

dose is below 1% (range: (-4.0 – +2.0)%) independently on the

plan type (table 1).

As shown in figure 1 the average pass rate obtained from the

array measurements is in good agreement (average

difference: (0.4 ± 1.0)%) with the average pass rate of the QA

measurements provided by the participating institutes

performed with their equipment for all the plans except for

the simple VMAT plan.

For the latter, the pass rate obtained with the Octavius is

influenced by the sensitivity variation of the array as a

function of gantry angle. Seven institutes out of 21 had plans

that failed the audit gamma analysis pass rate tolerance of

95% while the institute’s QA outcome was within tolerance (1

institute two plans, 6 institutes one plan). The film

measurement results are still under investigation and

therefore not presented in this abstract.

Conclusion:

The results demonstrate that such a national QA

audit is feasible. The reported in-house QA results were

consistent with the audit despite differences in dosimetry

equipment and analysis methods. Of the 21 Dutch centres

audited, 67% passed the gamma analysis test for all the plans

measured with a 2D-array by the audit team showing

acceptable implementation of IMRT and VMAT delivery.

OC-0546

The development of proton-beam grid therapy (PBGT)

T. Henry

Stockholm University, Department of Medical Radiation

Physics, Stockholm, Sweden

, A. Valdman

, A. Siegbahn

Karolinska Institutet, Department of Oncology and

Pathology, Stockholm, Sweden

Purpose or Objective:

Radiotherapy with grids has previously

been carried out with photon beams. The grid method is used

as an attempt to exploit the clinical finding that normal

tissue can tolerate higher doses as the irradiated volumes

become smaller. In this work we investigated the possibilities

to perform proton-beam grid therapy (PBGT) with millimeter-

wide proton beams by performing Monte Carlo simulations of

dose distributions produced by such grids. We also prepared

proton-grid treatment plans with a TPS, using real patient

data and beam settings available at modern proton therapy

centers.

Material and Methods:

Monte Carlo calculations were

performed using TOPAS version 1.2.p2 in a 20x20x20 cm3

water tank. The beam grids (each containing 4x4 proton

beams arranged in a square matrix) were aimed towards a

cubic target at the tank center. A total of 2x2 opposing grid

angles were used. The target was cross-fired in an interlaced

manner. A beam-size study was carried out to find a suitable

elemental beam size regarding beam thinness, peak-to-

entrance dose ratio and lateral penumbra along the beam

path. Dose distributions inside and outside of the target were

calculated for beam center-to-center (c-t-c) separations

inside the grids of 6, 8 and 10 mm.

The TPS study was performed with Varian Eclipse. We re-

planned two patients (one liver cancer and one rectal cancer

patient) already treated in the hospital with photon therapy

with the suggested PBGT. The IMPT method was used to

prepare these plans. The plan objectives were set to create a

homogeneous dose inside the target.

Results:

A beam size of 3 mm (FWHM) at the tank surface

was found suitable from a dosimetric point of view for the

further studies. By interlacing simulated beam grids from

several directions, a cubic and nearly homogeneous dose

distribution could be achieved in the target (see Figure 1).

The c-t-c distance was found to have a significant impact on

the valley dose outside of the target and on the homogeneity

of the target dose. In the TPS study, a rather uniform dose

distribution could be obtained inside of the contoured PTV

while preserving the grid pattern of the dose distribution

outside of it. The latter finding could be important for tissue

repair and recovery.